IBM Starts Rollout of 'watsonx' for Enterprise AI -- Virtualization Review
IBM Launches 'WatsonX' For Enterprise AI
IBM, a former leader in the realm of "cutting-edge" artificial intelligence, may have lagged behind in the inventive AI field, but is now striving to close the gap with a corporate solution named watsonx.
The blog section expands on the company's Watson AI platform, which gained recognition when an IBM supercomputer named Watson outperformed humans in the popular TV game show "Jeopardy!" back in 2011. In addition, the company is renowned for developing Deep Blue, a supercomputer that achieved excellence in chess thanks to advanced machine learning algorithms.
However, IBM seems to be falling behind in the latest competition of generative machine language initiated by ChatGPT, but it has recently taken steps to bridge the gap.
The official IBM WatsonX website states that the AI/data platform is comprised of three components:
Currently, the initial couple are accessible, but the widespread availability of watsonx.governance is anticipated in November.
IBM announced watsonx in May and has now shared that over 150 users from various industries have contributed to its development through the beta and tech preview programs.
According to IBM, individuals creating AI in watsonx.ai now have the opportunity to utilize models from both IBM and the Hugging Face community for various AI development activities. These models have already been trained to assist with different Natural Language Processing (NLP) tasks such as answering questions, generating and summarizing content, classifying text, and extracting information. In the upcoming updates, more IBM-trained foundational models will be made available, allowing for more efficient specialization in specific domains and tasks.
In addition to the basic models, watsonx.ai also offers a Prompt Lab for optimizing queries to large language models (LLMs). They will soon introduce a Tuning Studio that helps organizations enhance the performance and accuracy of their basic models by using labeled data.
IBM stated that watsonx will undergo further development in the coming year and they anticipate making significant updates. The main goal will be to extend the applications of the enterprise foundation model beyond natural language processing (NLP) and implement 100B+ parameter models for specific, customized purposes. This endeavor will pave the way for wider adoption of watsonx in various enterprises.
In addition, we will leverage our expertise in overseeing the use of artificial intelligence (AI) – assisting businesses in establishing comprehensive governance throughout the entire life cycle, reducing potential risks, and ensuring adherence to the increasing regulations in the AI and industry field. It is crucial to prioritize the implementation of AI governance, therefore, we advise our clients to initiate the governance of their machine learning (ML) models and core models right from the start.
The fresh product is obtainable for a trial without any charges before the pricing is implemented.
David Ramel works as an editor and writer at Converge360.
Please enter the characters you observe in the box above.
